There has been considerable public effort to identify, understand and eliminate racial and ethnic disparities in health and economic opportunities in the United States. A body of work has demonstrated the presence of wide disparities in population health across different subsets of the population, with racial and ethnic minorities subject to significantly higher morbidity and mortality from chronic disease on average. In addition to disparities in health, there is also widespread evidence documenting disparities in economic opportunity, with racial and ethnic minorities consistently earning lower pay for comparable work. While there are well-established disparities in health and earnings in minority populations, it is less understood how different sections of the population are affected by accidents in the workplace. Work-related injuries and disabilities impose large costs on society, and often have significant and lasting impacts on labor market outcomes for injured workers. Past studies have considered how the frequency and severity of occupational injuries differs across different racial and ethnic groups, but with relatively few definitive conclusions. In particular, it is largely unknown how the economic outcomes of injuries differ according to race. Conceptually, this is a difficult question to answer. On the one hand, the fact that most minorities have fewer employment opportunities means that they are less likely to work and thus face less exposure to workplace injury risk. However, when they do work they tend to work in riskier jobs, increasing their potential risk. Moreover, the effect of injuries on economic outcomes is complicated by their baseline disparities in employment and health. More work is needed to understand racial differences in economic outcomes after a workplace injury. Our study would address this issue and provide new evidence about racial and ethnic disparities in employment and earnings after a work-related disability. We will use data from a large, national survey with detailed information on employment, earnings, demographic characteristics and disability. We will use instrumental variables to estimate racial disparities in the frequency and severity of injuries independent of measurement error or selection bias. Finally, we will conduct statistical tests to identify how much of these disparities can be explained by observed characteristics and how much is unexplained and due to other factors such as discrimination.